[PSUBS-MAILIST] heads

[Sean T. Stevenson via Personal_Submersibles]

I presume that the achievable diametral tolerance is a result of the head thinning at the apex during hot pressing / spinning as material is pushed towards the flange. I forget the numbers now, but some time ago I posted a synopsis sourced from someone at EE explaining exactly what the typical thinning was. In any case, full hemispheres exhibit more thinning than 2:1 SE heads or dished heads of lesser depth, just because they are deeper. Two solutions to this are to form thicker than you need and then machine to remove the extraneous material away from the apex (expensive), or to form multiple spherically dished heads of shallower form and weld them together to fabricate your sphere. Nuytco uses six such dished heads on the DeepWorker spheres, pressed once on center and then a few times around the outside to get the best sphericity of the dish before trimming to shape and welding together in a jig to create the sphere. This corresponds to a cubic face construction, but you can do
this with with as few as four sections, corresponding to faces on a tetrahedron, or as many as twenty sections, corresponding to faces on an icosahedron.

Machining two hemispheres down to the apex thickness all over would give you the most theoretically perfect hull, but it is ridiculously expensive to do. Phil's method is a good tradeoff, at the expense of three times the welding.

Sean


On June 30, 2017 2:59:26 PM PDT, hank pronk via Personal_Submersibles <personal_submersibles at psubs.org> wrote:
>Hi Sean,I just heard back from EE with a cost for two 36 inch id  3.5
>inch heads, and they can not do 1% Diameter sphericity.  They can do
>ASME code witch is 1.25% or 5\8 inch.  I suppose that would mean
>machining the heads.  The heads could be machined to a very close
>tolerance then.  I would have to build a turret lathe then.Hank
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.whoweb.com/pipermail/personal_submersibles/attachments/20170630/248e5fec/attachment.html>